Obesity is a burgeoning public health problem with the rates of this condition differing by age, sex and ethnicity. The distribution of abdominal fat also differs by ethnic group. For example, compared to non-Hispanic Whites, African Americans have lower levels of visceral fat while having higher body mass index and waist circumference. These differences have important implications for the interpretation of the ethnic-specific relationships between abdominal body composition, inflammation and incident cardiovascular disease events. Notably, the standard anthropometric measure of body mass index cannot discriminate between different kinds or ratios of tissue types comprising the body weight. Accordingly, we propose to evaluate volumetric abdominal body composition in 1,970 men and women aged 45-84 at baseline, who are being studied in an ancillary study to the Multi-Ethnic Study of Atherosclerosis (MESA) for abdominal aortic calcium (AAC) by computed tomography (CT). The MESA has an extensive database of biomarkers and subclinical CVD measures on each participant, who are also being followed for incident CVD events. We propose to measure and calculate 5 discrete volumetric measures of abdominal body composition from the CT scans: subcutaneous fat, visceral fat, intramuscular fat, the visceral fat to visceral cavity ratio and the lean muscle to visceral fat ratio. We will also perform assays for several pertinent measures of inflammation. Using this and existing MESA data, we plan to test the hypothesis that measures of abdominal body composition will be significantly associated with incident coronary and cardiovascular disease events, independent of traditional CVD risk factors, inflammation and subclinical CVD. We will determine if there are differences in this association by age, sex and ethnicity and will also ascertain the influence of selected dietary and physical activity variables. Furthermore, the richness of the augmented MESA dataset will afford the opportunity to discern whether there is clear advantage afforded by the 5 distinct CT measures of body composition over the traditional measures of BMI and WC. This will provide important physiologic and clinically relevant information on the potential therapeutic targets for interventions aimed at reducing CVD risk by altering the composition of fat in different compartments, including skeletal muscle. We are unaware of any other population-based study that is as comprehensive as this to study the stated aims.